Data Cleaning: Removing NaN Values and Duplicates

Initial Data Exploration

import numpy as np

import pandas as pd

# Show numeric output in decimal format with 2 decimal places e.g., 2.15
pd.options.display.float_format = '{:,.2f}'.format

df_apps = pd.read_csv(r"C:\Users\4ben\Downloads\Google+Play+Store+Project+(Start)\apps.csv")
df_apps

	App	Category	Rating	Reviews	Size_MBs	Installs	Type	Price	Content_Rating	Genres	Last_Updated	Android_Ver
0	Ak Parti Yardım Toplama	SOCIAL	NaN	0	8.70	0	Paid	$13.99	Teen	Social	July 28, 2017	4.1 and up
1	Ain Arabic Kids Alif Ba ta	FAMILY	NaN	0	33.00	0	Paid	$2.99	Everyone	Education	April 15, 2016	3.0 and up
2	Popsicle Launcher for Android P 9.0 launcher	PERSONALIZATION	NaN	0	5.50	0	Paid	$1.49	Everyone	Personalization	July 11, 2018	4.2 and up
3	Command & Conquer: Rivals	FAMILY	NaN	0	19.00	0	NaN	0	Everyone 10+	Strategy	June 28, 2018	Varies with device
4	CX Network	BUSINESS	NaN	0	10.00	0	Free	0	Everyone	Business	August 6, 2018	4.1 and up
...	...	...	...	...	...	...	...	...	...	...	...	...
10836	Subway Surfers	GAME	4.50	27723193	76.00	1,000,000,000	Free	0	Everyone 10+	Arcade	July 12, 2018	4.1 and up
10837	Subway Surfers	GAME	4.50	27724094	76.00	1,000,000,000	Free	0	Everyone 10+	Arcade	July 12, 2018	4.1 and up
10838	Subway Surfers	GAME	4.50	27725352	76.00	1,000,000,000	Free	0	Everyone 10+	Arcade	July 12, 2018	4.1 and up
10839	Subway Surfers	GAME	4.50	27725352	76.00	1,000,000,000	Free	0	Everyone 10+	Arcade	July 12, 2018	4.1 and up
10840	Subway Surfers	GAME	4.50	27711703	76.00	1,000,000,000	Free	0	Everyone 10+	Arcade	July 12, 2018	4.1 and up

10841 rows × 12 columns

#Shape tells us the shape of an array.
#10841 rows and 12 columns.
df_apps.shape

(10841, 12)

#We can already see that there are some data cleansing activities we need to be executed.
#In the Ratings and Type columns there are NaN (Not a number values) and in the Price column we have dollar signs in our data entries which will cause calculation errors.

# sample(n) method
#We can use sample to give us a number of rows
df_apps.sample(10)

	App	Category	Rating	Reviews	Size_MBs	Installs	Type	Price	Content_Rating	Genres	Last_Updated	Android_Ver
8571	Real Madrid App	SPORTS	4.60	161423	73.00	5,000,000	Free	0	Everyone	Sports	August 1, 2018	5.0 and up
3409	Account Class-12 Solutions (D K Goel) Vol-2	FAMILY	4.60	124	23.00	10,000	Free	0	Everyone	Education	April 17, 2018	3.0 and up
36	D+H Reaction Wall	GAME	NaN	0	36.00	1	Paid	$0.99	Everyone	Arcade	July 24, 2016	Varies with device
3952	Swift Dark CM / CM13 Theme	PERSONALIZATION	4.70	1500	41.00	10,000	Paid	$1.99	Everyone	Personalization	January 5, 2017	5.0 and up
4996	FlashLight F.Light	TOOLS	4.30	3943	11.00	100,000	Free	0	Everyone	Tools	May 23, 2018	5.0 and up
1339	PCOS Guide - Fight PCOS naturally	MEDICAL	NaN	0	5.10	100	Free	0	Everyone	Medical	August 4, 2018	5.0 and up
8350	American Airlines	TRAVEL_AND_LOCAL	3.70	16973	12.00	5,000,000	Free	0	Everyone	Travel & Local	July 18, 2018	Varies with device
6007	Pro App for Craigslist	SHOPPING	4.10	5618	4.60	500,000	Free	0	Teen	Shopping	May 9, 2018	4.1 and up
3662	Access Point Names	COMMUNICATION	4.00	138	2.40	10,000	Free	0	Everyone	Communication	August 12, 2014	2.3.3 and up
5942	Wheretoget: Shop in style	LIFESTYLE	4.10	6808	12.00	500,000	Free	0	Teen	Lifestyle	June 5, 2017	4.1 and up

Removing redundant columns and NaN Values

#To remove the unwanted columns, we simply provide a list of the column names ['Last_Updated', ‘Android_Ver'] to the .drop() method. 
#By setting axis=1 we are specifying that we want to drop certain columns.
#inplace function permanently modifies the current dataframe in place 

df_apps.drop(["Last_Updated","Android_Ver"],axis=1 , inplace=True)
df_apps.sample(10)

	App	Category	Rating	Reviews	Size_MBs	Installs	Type	Price	Content_Rating	Genres
6430	Used car is the first car - used car purchase,...	AUTO_AND_VEHICLES	4.60	5097	19.00	1,000,000	Free	0	Everyone	Auto & Vehicles
91	4 Paws PH	MEDICAL	NaN	1	29.00	5	Free	0	Everyone	Medical
8331	Univision Deportes: Liga MX, MLS, Fútbol Live	SPORTS	4.20	75545	14.00	5,000,000	Free	0	Everyone	Sports
10155	trivago: Hotels & Travel	TRAVEL_AND_LOCAL	4.20	219848	12.00	50,000,000	Free	0	Everyone	Travel & Local
6507	CZ File Manager	TOOLS	2.20	876	8.70	1,000,000	Free	0	Everyone	Tools
8717	Ginger Keyboard - Emoji, GIFs, Themes & Games	PRODUCTIVITY	4.40	162831	43.00	5,000,000	Free	0	Everyone	Productivity
2594	Sam.BN	TOOLS	NaN	83	2.00	1,000	Free	0	Everyone	Tools
7227	Car Driving Simulator Drift	GAME	4.40	19816	57.00	1,000,000	Free	0	Everyone	Racing
7237	Discovery K!ds Play!	FAMILY	4.10	19388	35.00	1,000,000	Free	0	Everyone	Entertainment;Music & Video
2323	BP-Tech	FAMILY	3.20	12	13.00	1,000	Free	0	Everyone	Casual

To find and remove the rows with the NaN (not a number) values we can create a subset of the DataFrame based on where .isna() evaluates to True. We see that NaN values in ratings are associated with no reviews (and no installs), which makes sense.

df_apps_clean = df_apps.dropna() #Drops the NaN values
df_apps_clean.shape # We now have 9365 rows after removing the rows which had NaN values.

(9367, 10)

#Double checking for any nan values and we see a "False" result which means there are no more NaN entries. 
c = df_apps_clean.isna().values.any()
c

False

Look for duplicate rows

duplicated_rows = df_apps_clean[df_apps_clean.duplicated()]
print(duplicated_rows.shape) # We use print to return a value in combination with other lines of code in a cell. 
duplicated_rows.head(10)

(476, 10)

	App	Category	Rating	Reviews	Size_MBs	Installs	Type	Price	Content_Rating	Genres
946	420 BZ Budeze Delivery	MEDICAL	5.00	2	11.00	100	Free	0	Mature 17+	Medical
1133	MouseMingle	DATING	2.70	3	3.90	100	Free	0	Mature 17+	Dating
1196	Cardiac diagnosis (heart rate, arrhythmia)	MEDICAL	4.40	8	6.50	100	Paid	$12.99	Everyone	Medical
1231	Sway Medical	MEDICAL	5.00	3	22.00	100	Free	0	Everyone	Medical
1247	Chat Kids - Chat Room For Kids	DATING	4.70	6	4.90	100	Free	0	Mature 17+	Dating
1379	CT Scan Cross Sectional Anatomy	MEDICAL	4.30	10	46.00	100	Free	0	Everyone	Medical
1616	JH Blood Pressure Monitor	MEDICAL	3.70	9	2.90	500	Free	0	Everyone	Medical
1642	Cardi B Live Stream Video Chat - Prank	DATING	4.40	28	3.40	500	Free	0	Everyone	Dating
1813	Diabetes & Diet Tracker	MEDICAL	4.60	395	19.00	1,000	Paid	$9.99	Everyone	Medical
1821	Transenger – Ts Dating and Chat for Free	DATING	3.60	8	14.00	1,000	Free	0	Mature 17+	Dating

duplicated_rows.shape

(476, 10)

duplicated_rows.head()

	App	Category	Rating	Reviews	Size_MBs	Installs	Type	Price	Content_Rating	Genres
946	420 BZ Budeze Delivery	MEDICAL	5.00	2	11.00	100	Free	0	Mature 17+	Medical
1133	MouseMingle	DATING	2.70	3	3.90	100	Free	0	Mature 17+	Dating
1196	Cardiac diagnosis (heart rate, arrhythmia)	MEDICAL	4.40	8	6.50	100	Paid	$12.99	Everyone	Medical
1231	Sway Medical	MEDICAL	5.00	3	22.00	100	Free	0	Everyone	Medical
1247	Chat Kids - Chat Room For Kids	DATING	4.70	6	4.90	100	Free	0	Mature 17+	Dating

df_apps_clean.duplicated()

21       False
28       False
47       False
82       False
99       False
         ...  
10836    False
10837    False
10838    False
10839     True
10840    False
Length: 9367, dtype: bool

df_apps_clean.duplicated().sum() 
#476 duplicate rows

476

We need to provide the column names that should be used in the comparison to identify duplicates.

df_apps_clean = df_apps_clean.drop_duplicates(subset=["App","Type","Price"]) #Drops the rows where there are duplicates based on App,Type and Price column.
df_apps_clean.shape

(8199, 10)

Preliminary Data Exploration

#Sort apps based on the highest rating
df_apps_clean.sort_values('Rating', ascending=False).head(9)

	App	Category	Rating	Reviews	Size_MBs	Installs	Type	Price	Content_Rating	Genres
21	KBA-EZ Health Guide	MEDICAL	5.00	4	25.00	1	Free	0	Everyone	Medical
1230	Sway Medical	MEDICAL	5.00	3	22.00	100	Free	0	Everyone	Medical
1227	AJ Men's Grooming	LIFESTYLE	5.00	2	22.00	100	Free	0	Everyone	Lifestyle
1224	FK Dedinje BGD	SPORTS	5.00	36	2.60	100	Free	0	Everyone	Sports
1223	CB VIDEO VISION	PHOTOGRAPHY	5.00	13	2.60	100	Free	0	Everyone	Photography
1222	Beacon Baptist Jupiter, FL	LIFESTYLE	5.00	14	2.60	100	Free	0	Everyone	Lifestyle
1214	BV Mobile Apps	PRODUCTIVITY	5.00	3	4.80	100	Free	0	Everyone	Productivity
2680	Florida Wildflowers	FAMILY	5.00	5	69.00	1,000	Free	0	Everyone	Education
1206	ADS-B Driver	TOOLS	5.00	2	6.30	100	Paid	$1.99	Everyone	Tools

#Sort apps based on the Size_MBs (Highest to Lowest)
df_apps_clean.sort_values('Size_MBs', ascending=False).head(9)

	App	Category	Rating	Reviews	Size_MBs	Installs	Type	Price	Content_Rating	Genres
9942	Talking Babsy Baby: Baby Games	LIFESTYLE	4.00	140995	100.00	10,000,000	Free	0	Everyone	Lifestyle;Pretend Play
10687	Hungry Shark Evolution	GAME	4.50	6074334	100.00	100,000,000	Free	0	Teen	Arcade
9943	Miami crime simulator	GAME	4.00	254518	100.00	10,000,000	Free	0	Mature 17+	Action
9944	Gangster Town: Vice District	FAMILY	4.30	65146	100.00	10,000,000	Free	0	Mature 17+	Simulation
3144	Vi Trainer	HEALTH_AND_FITNESS	3.60	124	100.00	5,000	Free	0	Everyone	Health & Fitness
9945	Ultimate Tennis	SPORTS	4.30	183004	100.00	10,000,000	Free	0	Everyone	Sports
7926	Post Bank	FINANCE	4.50	60449	100.00	1,000,000	Free	0	Everyone	Finance
7927	The Walking Dead: Our World	GAME	4.00	22435	100.00	1,000,000	Free	0	Teen	Action
7928	Stickman Legends: Shadow Wars	GAME	4.40	38419	100.00	1,000,000	Paid	$0.99	Everyone 10+	Action

#Sort apps based on the Reviews (Highest to Lowest)
df_apps_clean.sort_values('Reviews', ascending=False).head(20)

	App	Category	Rating	Reviews	Size_MBs	Installs	Type	Price	Content_Rating	Genres
10805	Facebook	SOCIAL	4.10	78158306	5.30	1,000,000,000	Free	0	Teen	Social
10785	WhatsApp Messenger	COMMUNICATION	4.40	69119316	3.50	1,000,000,000	Free	0	Everyone	Communication
10806	Instagram	SOCIAL	4.50	66577313	5.30	1,000,000,000	Free	0	Teen	Social
10784	Messenger – Text and Video Chat for Free	COMMUNICATION	4.00	56642847	3.50	1,000,000,000	Free	0	Everyone	Communication
10650	Clash of Clans	GAME	4.60	44891723	98.00	100,000,000	Free	0	Everyone 10+	Strategy
10744	Clean Master- Space Cleaner & Antivirus	TOOLS	4.70	42916526	3.40	500,000,000	Free	0	Everyone	Tools
10835	Subway Surfers	GAME	4.50	27722264	76.00	1,000,000,000	Free	0	Everyone 10+	Arcade
10828	YouTube	VIDEO_PLAYERS	4.30	25655305	4.65	1,000,000,000	Free	0	Teen	Video Players & Editors
10746	Security Master - Antivirus, VPN, AppLock, Boo...	TOOLS	4.70	24900999	3.40	500,000,000	Free	0	Everyone	Tools
10584	Clash Royale	GAME	4.60	23133508	97.00	100,000,000	Free	0	Everyone 10+	Strategy
10763	Candy Crush Saga	GAME	4.40	22426677	74.00	500,000,000	Free	0	Everyone	Casual
10770	UC Browser - Fast Download Private & Secure	COMMUNICATION	4.50	17712922	40.00	500,000,000	Free	0	Teen	Communication
10735	Snapchat	SOCIAL	4.00	17014787	5.30	500,000,000	Free	0	Teen	Social
10489	360 Security - Free Antivirus, Booster, Cleaner	TOOLS	4.60	16771865	3.40	100,000,000	Free	0	Everyone	Tools
10731	My Talking Tom	GAME	4.50	14891223	36.00	500,000,000	Free	0	Everyone	Casual
10594	8 Ball Pool	GAME	4.50	14198297	52.00	100,000,000	Free	0	Everyone	Sports
10302	DU Battery Saver - Battery Charger & Battery Life	TOOLS	4.50	13479633	14.00	100,000,000	Free	0	Everyone	Tools
10354	BBM - Free Calls & Messages	COMMUNICATION	4.30	12842860	3.50	100,000,000	Free	0	Everyone	Communication
10549	Cache Cleaner-DU Speed Booster (booster & clea...	TOOLS	4.50	12759663	15.00	100,000,000	Free	0	Everyone	Tools
10757	Twitter	NEWS_AND_MAGAZINES	4.30	11667403	6.30	500,000,000	Free	0	Mature 17+	News & Magazines

Create Pie and Donut Charts

We will use: - plotly - a commonly used data visualisation library that you can use in combination with or instead of Matplotlib.

#Counting the number of entries from each Content_rating
ratings = df_apps_clean.Content_Rating.value_counts()
ratings

Everyone           6621
Teen                912
Mature 17+          357
Everyone 10+        305
Adults only 18+       3
Unrated               1
Name: Content_Rating, dtype: int64

#Importing the data visualisation libraries
import pandas as pd
import plotly.express as px

#The px.pie function is called to create the pie chart. We input our labels and values into the variable (fig)

fig = px.pie(labels=ratings.index,values=ratings.values,title="Content Rating",names=ratings.index,)

fig.update_traces(textposition='outside', textinfo='percent+label')

fig.show()

If you’d like to configure other aspects of the chart, that you can’t see in the list of parameters, you can call a method called .update_traces(). In plotly lingo, “traces” refer to graphical marks on a figure. Think of “traces” as collections of attributes. Here we update the traces to change how the text is displayed.

To create a donut 🍩 chart

fig = px.pie(labels=ratings.index,values=ratings.values,title="Content Rating",names=ratings.index,hole=0.6,)

fig.update_traces(textposition='inside', textfont_size=15, textinfo='percent')
 
fig.show()

Modifying the datatype of particular columns

#Installs data type is not coming up as a number. But rather coming up as an object.
#The reason Python is not recognising our installs as numbers is because of the comma (,) characters in the Installs column.
df_apps_clean.Installs.describe()

count        8199
unique         19
top       1000000
freq         1417
Name: Installs, dtype: object

We can remove the comma (,) character - or any character for that matter - from a DataFrame using the string’s .replace() method. Here we’re saying: “replace the , with an empty string”. This completely removes all the commas in the Installs column. We can then convert our data to a number using .to_numeric().

df_apps_clean.Installs = df_apps_clean.Installs.astype(str).str.replace(',', "")  
#removes the comma from the numbers in the installs column and replaces it with an empty/blank string.

df_apps_clean.Installs = pd.to_numeric(df_apps_clean.Installs) 
#convert the data in Installs column into a number

#The price is showing as an object rather than a number data type, so we need to change it.

df_apps_clean.Price.describe()

count     8199
unique      73
top          0
freq      7595
Name: Price, dtype: object

#We use replace method to filter out the dollar sign from all the entries in the Price column.
df_apps_clean.Price = df_apps_clean.Price.astype(str).str.replace('$', "")

C:\Users\4ben\AppData\Local\Temp/ipykernel_17652/2313303643.py:2: FutureWarning:

The default value of regex will change from True to False in a future version. In addition, single character regular expressions will *not* be treated as literal strings when regex=True.

#After removing the dollar sign from the entries in the Price column, we then change price column to a number or float data type

df_apps_clean.Price = pd.to_numeric(df_apps_clean.Price)

df_apps_clean.Price.describe()

count   8,199.00
mean        1.04
std        16.85
min         0.00
25%         0.00
50%         0.00
75%         0.00
max       400.00
Name: Price, dtype: float64

df_apps_clean
# We can see that the dollar signs no longer appear in the data entries within the Price column. 

	App	Category	Rating	Reviews	Size_MBs	Installs	Type	Price	Content_Rating	Genres
21	KBA-EZ Health Guide	MEDICAL	5.00	4	25.00	1	Free	0.00	Everyone	Medical
28	Ra Ga Ba	GAME	5.00	2	20.00	1	Paid	1.49	Everyone	Arcade
47	Mu.F.O.	GAME	5.00	2	16.00	1	Paid	0.99	Everyone	Arcade
82	Brick Breaker BR	GAME	5.00	7	19.00	5	Free	0.00	Everyone	Arcade
99	Anatomy & Physiology Vocabulary Exam Review App	MEDICAL	5.00	1	4.60	5	Free	0.00	Everyone	Medical
...	...	...	...	...	...	...	...	...	...	...
10824	Google Drive	PRODUCTIVITY	4.40	2731171	4.00	1000000000	Free	0.00	Everyone	Productivity
10828	YouTube	VIDEO_PLAYERS	4.30	25655305	4.65	1000000000	Free	0.00	Teen	Video Players & Editors
10829	Google Play Movies & TV	VIDEO_PLAYERS	3.70	906384	4.65	1000000000	Free	0.00	Teen	Video Players & Editors
10831	Google News	NEWS_AND_MAGAZINES	3.90	877635	13.00	1000000000	Free	0.00	Teen	News & Magazines
10835	Subway Surfers	GAME	4.50	27722264	76.00	1000000000	Free	0.00	Everyone 10+	Arcade

8199 rows × 10 columns

df_apps_clean.sort_values('Price', ascending=False).head(20)

	App	Category	Rating	Reviews	Size_MBs	Installs	Type	Price	Content_Rating	Genres
3946	I'm Rich - Trump Edition	LIFESTYLE	3.60	275	7.30	10000	Paid	400.00	Everyone	Lifestyle
2461	I AM RICH PRO PLUS	FINANCE	4.00	36	41.00	1000	Paid	399.99	Everyone	Finance
4606	I Am Rich Premium	FINANCE	4.10	1867	4.70	50000	Paid	399.99	Everyone	Finance
3145	I am rich(premium)	FINANCE	3.50	472	0.94	5000	Paid	399.99	Everyone	Finance
3554	💎 I'm rich	LIFESTYLE	3.80	718	26.00	10000	Paid	399.99	Everyone	Lifestyle
5765	I am rich	LIFESTYLE	3.80	3547	1.80	100000	Paid	399.99	Everyone	Lifestyle
1946	I am rich (Most expensive app)	FINANCE	4.10	129	2.70	1000	Paid	399.99	Teen	Finance
2775	I Am Rich Pro	FAMILY	4.40	201	2.70	5000	Paid	399.99	Everyone	Entertainment
3221	I am Rich Plus	FAMILY	4.00	856	8.70	10000	Paid	399.99	Everyone	Entertainment
3114	I am Rich	FINANCE	4.30	180	3.80	5000	Paid	399.99	Everyone	Finance
1331	most expensive app (H)	FAMILY	4.30	6	1.50	100	Paid	399.99	Everyone	Entertainment
2394	I am Rich!	FINANCE	3.80	93	22.00	1000	Paid	399.99	Everyone	Finance
3897	I Am Rich	FAMILY	3.60	217	4.90	10000	Paid	389.99	Everyone	Entertainment
2193	I am extremely Rich	LIFESTYLE	2.90	41	2.90	1000	Paid	379.99	Everyone	Lifestyle
3856	I am rich VIP	LIFESTYLE	3.80	411	2.60	10000	Paid	299.99	Everyone	Lifestyle
2281	Vargo Anesthesia Mega App	MEDICAL	4.60	92	32.00	1000	Paid	79.99	Everyone	Medical
1407	LTC AS Legal	MEDICAL	4.00	6	1.30	100	Paid	39.99	Everyone	Medical
2629	I am Rich Person	LIFESTYLE	4.20	134	1.80	1000	Paid	37.99	Everyone	Lifestyle
2481	A Manual of Acupuncture	MEDICAL	3.50	214	68.00	1000	Paid	33.99	Everyone	Medical
4264	Golfshot Plus: Golf GPS	SPORTS	4.10	3387	25.00	50000	Paid	29.99	Everyone	Sports

Find number of unique items in a column

df_apps_clean.Category.nunique()
#33 unique categories

33

Number of apps per category

top10_category = df_apps_clean.Category.value_counts()[:10] # Entries from index number 0 to 10 are returned.
print(top10_category)
#index values are on left column and values are on right column

FAMILY             1610
GAME                910
TOOLS               719
FINANCE             302
LIFESTYLE           302
PRODUCTIVITY        301
PERSONALIZATION     298
MEDICAL             292
PHOTOGRAPHY         263
BUSINESS            262
Name: Category, dtype: int64

Bar Chart

bar = px.bar(x = top10_category.index, # index = category name
             y = top10_category.values)
 
bar.show()

How many installs has each category had

#Group all our apps by category and sum the number of installations that each category has had.
category_installs = df_apps_clean.groupby('Category').agg({'Installs': pd.Series.sum})
category_installs.sort_values('Installs', ascending=True, inplace=True)

category_installs

	Installs
Category
EVENTS	15949410
BEAUTY	26916200
PARENTING	31116110
MEDICAL	39162676
COMICS	44931100
LIBRARIES_AND_DEMO	52083000
AUTO_AND_VEHICLES	53129800
HOUSE_AND_HOME	97082000
ART_AND_DESIGN	114233100
DATING	140912410
FOOD_AND_DRINK	211677750
EDUCATION	352852000
WEATHER	361096500
FINANCE	455312400
MAPS_AND_NAVIGATION	503267560
LIFESTYLE	503742120
BUSINESS	692018120
SPORTS	1096431465
HEALTH_AND_FITNESS	1134006220
SHOPPING	1400331540
PERSONALIZATION	1532352930
BOOKS_AND_REFERENCE	1665791655
ENTERTAINMENT	2113660000
NEWS_AND_MAGAZINES	2369110650
TRAVEL_AND_LOCAL	2894859300
VIDEO_PLAYERS	3916897200
FAMILY	4437579590
PHOTOGRAPHY	4649143130
SOCIAL	5487841475
PRODUCTIVITY	5788070180
TOOLS	8099724500
COMMUNICATION	11039241530
GAME	13858762717

Grouped Bar Charts & Box Plots

df_free_vs_paid = df_apps_clean.groupby(["Category", "Type"], as_index=False).agg({'App': pd.Series.count})
df_free_vs_paid.head()

	Category	Type	App
0	ART_AND_DESIGN	Free	58
1	ART_AND_DESIGN	Paid	3
2	AUTO_AND_VEHICLES	Free	72
3	AUTO_AND_VEHICLES	Paid	1
4	BEAUTY	Free	42

g_bar = px.bar(df_free_vs_paid,
               x='Category',
               y='App',
               title='Free vs Paid Apps by Category',
               color='Type',
               barmode='group')
 
g_bar.update_layout(xaxis_title='Category',
                    yaxis_title='Number of Apps',
                    xaxis={'categoryorder':'total descending'},
                    yaxis=dict(type='log'))
 
g_bar.show()